246 H. S. Harrison, 



and we have to consider two chief possibilities — either the spaces were 

 the first to appear, and have been the direct cause of the formation 

 of the apertures, 07' the association between spaces and apertures is 

 accidental. I shall attempt to show that the arguments are a priori 

 and a posteriori in favour of the former view, and that by accepting 

 it we not only violate no physiological or morphological probabilities, 

 but also obtain a clearer conception of one of the most important steps 

 in the evolution of the auditory organ of the Vertebrata. 



At the beginning of his account of the perilymphatic spaces of 

 the internal ear of Vertebrates, Hasse [1] says, ". . . . ich wende mich 

 daher zum zweiten Teile meiner Aufgabe, zur Schilderung der Art und 

 Weise, auf welche die das Labyrinth umspülende Flüssigkeit, der liquor' 

 perilymphaticus, sich erneuern und ergänzen kann." In another paper 

 [3] he expresses the view that the perilymphatic spaces have exercised 

 considerable influence on the development of the various parts of the 

 labyrinth, owing to the advantages gained by the transmission of 

 variations of pressure through a fluid. We have here the suggestion 

 of what seem to be the only probable functions of the perilymphatic 

 spaces and the perilymph. The thinness of the wall separating endo- 

 lymph from perilymph over the tympanal areas is very suggestive of 

 the probability of an osmotic process for the transference of waste 

 products from one to the other — from endolymph to perilymph. As 

 regards the direct communication between the perilymphatic spaces and 

 the lymphatic system^), it seems probable that such an arrangement is 

 not only unnecessary for the circulation of the perilymph, but would 

 be prejudicial to the more important mechanical function of conveying 

 vibrations. These would be transmitted with greater efficiency in a 

 restricted system of fluid-containing spaces, than in one whose limits 

 were as indefinite as Hasse and Retzius believed to be the case with 

 the perilymphatic system. I believe the system to be closed and de- 

 finite. A renewal or purification of the perilymph can probably take 

 place with sufficient rapidity through the thin walls separating peri- 



^) r have found no cellular elements in the perilymphatic fluid. 



